July, 2009 Linda O’Brien-Pallas, PhD, RN Gail Tomblin Murphy, PhD, RN Judith Shamian, PhD, RN

Nursing Health Services Research Unit

Co-Principal Investigators

Co-Investigators 



Linda O’Brien-Pallas,

RN,PhD, FCAHS



email:[email protected]





Gail Tomblin Murphy,





Judith Shamian,

RN, PhD

RN, PhD, LLD

email: [email protected]

  

X. Mingyang Li, PhD George Kephart, PhD Heather Laschinger, PhD Marlene Smadu, EdD Linda McGillis-Hall, PhD Danielle D’Amour, PhD Mae Gallant, RN, MScN

Any reproduction requires permission from TOS co-PIs

2

Support from across Canada include:  1 primary funder  6 co-sponsors  19 decision makers  41 hospitals

Any reproduction requires permission from TOS co-PIs

3



To give a brief background on the Nursing Turnover Study (TOS).



To highlight overall findings from study.

Any reproduction requires permission from TOS co-PIs

4



Nursing Research over the last twenty years focused on: ◦ ◦ ◦ ◦



Nurses Patients Nurses and patients Nurses, patients, and work environments

This study focused on Nursing Turnover and the Cost Associated with Turnover.

Any reproduction requires permission from TOS co-PIs

5



To arrive at new evidence about the incidence of nurse turnover and its predictors, and to examine its impact on patient and nurse outcomes and associated system costs.



Inform policies to effectively retain and recruit nurses.



Provide nursing data from a Canadian context.

6



Loss of human capital as nurses leave and loss of productivity as new hires are oriented.

Any reproduction requires permission from TOS co-PIs

7

1.

2. 3. 4. 5.

What is the relationship between system inputs and nurse, patient and system outcomes? How do system inputs influence system throughputs? How does system throughput mediate both inputs and outputs? How do system outputs feed back into the system and what are the implications? How might modifications of throughputs alter system outcomes? Any reproduction requires permission from TOS co-PIs

9

Data Collection  Longitudinal  Two waves of data was collected on patients, nurses, units and hospitals.  Each wave consisted of a three-month data collection period staggered over one year (seasonal effect). 

Wave 1 occurred from Mar. to Nov. 2005



Wave 2 occurred from Jan. to Aug. 2006

Any reproduction requires permission from TOS co-PIs

10

Thirteen Study Instruments 1. Hospital Data  Hospital Profile 2. Unit Data  Unit Profile  Nurse Unit Variables  Financial Variables  Staffing Variables  Turnover Unit Cost  Environmental Complexity Scale (ECS) Any reproduction requires permission from TOS co-PIs

11

Study Instruments 3. Nurse  Nurse Survey  Nurse Outcome Variables  Reasons for Leaving  Turnover Vacancy Costs 4. Patient  Patient Judgment of Hospital Quality Survey  Patient Outcome Variables Any reproduction requires permission from TOS co-PIs

12

1.Descriptive analysis 2.HLM analysis 3.Simulation modeling

Any reproduction requires permission from TOS co-PIs

13

Level of Data Wave 1 Hospital

Wave 2

41

39

182

163

Nurse

4,481

3,844

Patient

4,412

3,726

Unit

Any reproduction requires permission from TOS co-PIs

14

Any reproduction requires permission from TOS co-PIs

15

One Year Turnover Rate 20.8%

16.4%

13.7%

26.7%

17.8% 19.1% 20.8% 19.8% 18.8%

Medical/Surgical Medical Paed hosp Psychiatric Surgical

ICU Obstetrics/Gynecology Paed-unit/dept w/i hosp Rehab/LTC/Geriatric Any reproduction requires permission from TOS co-PIs

16









Greater than 50% of the Turnover occurs in medical, surgical, and medical surgical units. One in two turnovers happens in a medical or surgical units. ICUs are of concern as there is 26.7% turnover –every fourth nurse in an ICU leaves per year. This is a system and patient safety concern ◦ .

Any reproduction requires permission from TOS co-PIs

17

Any reproduction requires permission from TOS co-PIs

18

$21,725 $23,121

Surgical

$15,201

Rehab/LTC/Geriatric

$28,767 $14,658 $16,225

Unit Type

Psychiatric

$64,606

Paed-unit/dept w/i hosp

$23,982 Wave 1

$20,610 $21,502

Paed hosp

Wave 2

$17,409

Obstetrics/Gynecology

$8,917 $21,164

Medical/Surgical

$35,199 $22,757 $28,404

Medical

$25,412

ICU

$35,841 0

10,000

20,000

30,000

40,000

Total Dollar

50,000

60,000

70,000 19

Direct cost was about $10,900 and $15,400 on average, waves 1 & 2 respectively.  Indirect cost was about $15,400 and $11,000 on average, waves 1 & 2 respectively. 

Any reproduction requires permission from TOS co-PIs

20

Any reproduction requires permission from TOS co-PIs

21

Any reproduction requires permission from TOS co-PIs

22

90% were RNs.  At least 40% had BScN.  Average age was 39 years old with on average 14 years of nursing experience.  60% of nurses were full time basis and 34% part time. 

Any reproduction requires permission from TOS co-PIs

23



Nurse Satisfaction



Nurse Leadership



Health of Nurses



Cost and Productivity of New Hires

Any reproduction requires permission from TOS co-PIs

24



Nurses who reported higher job satisfaction scores at wave 1 also reported high job satisfaction scores at wave 2.



RNs were less satisfied than RPNs/LPNs.



Nurses on units with better leadership were more satisfied.



Higher job satisfaction is associated with lower turnover rate on the unit. Any reproduction requires permission from TOS co-PIs

25







Better leadership is associated with better mental health and higher job satisfaction. Better leadership is associated with increased productivity on the unit Better leadership was associated with lower turnover rate on the unit. Any reproduction requires permission from TOS co-PIs

26











Over 2/3 of respondents reported missing work because of physical illness. More experienced nurses reported poorer physical health status but better mental health status. Less experienced nurses reported poorer mental health status and better physical status health.

Better leadership on the unit was associated with better mental health status. Higher turnover rate were associated with deterioration in nurse’s mental health status. Any reproduction requires permission from TOS co-PIs

27

New hires were higher educated, but less experienced than the existing nurses.  Average mean cost of decreased productivity was $17,800 (w1) and $8,500 (w2). 

Any reproduction requires permission from TOS co-PIs

28

Any reproduction requires permission from TOS co-PIs

29



62% of patient population sample came from surgical, medical and pediatric units.

Any reproduction requires permission from TOS co-PIs

30

2/3 of the patients experienced prolonged LOS that averaged 12 -15 days.  ICU, Medical and Med/Surg units encountered the most cases with serious to life threatening complexity; and reported the highest use of resources. 

Any reproduction requires permission from TOS co-PIs

31

Med/surgical and surgical units reported higher occurrence of medical errors than the overall average in both waves (greater than 10% in both waves).  Overall average was 7% and 3.4%, waves 1 and 2 respectively. 

Any reproduction requires permission from TOS co-PIs

32

Focus on relationship between and among patient, nurse, and system outcomes

Any reproduction requires permission from TOS co-PIs

33





To study Pulling factors at wave 1 attributed to nurses’ entry to a unit at wave 2. For units with more experienced new hires at wave 1, they are more likely to have new hires at wave 2.  Turnover in W1 was predictor of turnover in W2  This was a direct relationship and indicator of systemic issues.



For units with higher full time mix, they are less likely to have new hires.

Any reproduction requires permission from TOS co-PIs

35

 





Older new hires are more experienced. The RN new hires are more experienced than RPN and LPNs. A unit with higher productivity or higher environmental complexity is less likely to hire more experienced nurses (system issue). A unit with higher proportion of overtime is less likely to hire more experienced nurses (system issue). Any reproduction requires permission from TOS co-PIs

36





Nurses with better physical health at wave 1 had better physical health at wave 2. Better leadership on unit is associated with worse physical health.

Any reproduction requires permission from TOS co-PIs

37







Nurses with better mental health status at wave 1 had better mental health status at wave 2. Nurses with degree and better leadership on the unit have better mental health status. Increased role conflict, increased acuity of patients, and higher turnover rates result in poor mental health status for nurses.

Any reproduction requires permission from TOS co-PIs

38



In units with effective leadership and lower turnover rates, there was higher job satisfaction.

Any reproduction requires permission from TOS co-PIs

39







Probability of wave 2 patients having at least one medical error. Medical errors increase with acuity, role ambiguity, and increased turnover rates. Medical errors increase with poor leadership and reduced productivity of nurses.

Any reproduction requires permission from TOS co-PIs

40





Probability of death increases for wave 2 patient with one or more complications (UTI, shock, pneumonia, post-op infection, GI bleeds, cardiac/resp. arrests, undocumented). FTR increases with age of patients, in ICUs and medical units, and in environments with poor team dynamics.

Any reproduction requires permission from TOS co-PIs

41



Prolonged stay (actual versus expected) increased with acuity of patients and, on units with increased role conflict.

Any reproduction requires permission from TOS co-PIs

42





Turnover rates were higher on medical and medical surgical units and in ICUs, and on units with increased role ambiguity and role conflict, and increased use of overtime. Turnover rates were lower on units with increased full time mix.

Any reproduction requires permission from TOS co-PIs

43



Turnover is associated with:  decreased job satisfaction  increased likelihood of medical errors, overtime and environmental complexity.

Any reproduction requires permission from TOS co-PIs

45





Better leadership on the unit is associated with better mental health, higher job satisfaction and higher productivity. Non-supportive working environments and poor relationships with team members contribute to nurses’ decisions to leave. Any reproduction requires permission from TOS co-PIs

46





Role ambiguity and conflict on units are associated with higher turnover rate for nurses. Higher proportion of full time nurses is associated with lower nursing turnover rates.

Any reproduction requires permission from TOS co-PIs

47





Medical errors are related to higher levels of turnover and role ambiguity for nurses on the unit. Prolonged LOS was observed on units when role conflict level increased. Any reproduction requires permission from TOS co-PIs

48





Reliable comparison of turnover cost remains problematic due to the varying definition of unit of measures used by hospitals. Smaller than anticipated sample weakened ability to perform more robust analysis. Any reproduction requires permission from TOS co-PIs

49

Contributions of Findings to Simulation Modeling

Any reproduction requires permission from TOS co-PIs

50









Evidence from this study can be used to estimate mathematical relationships not previously known. This allows for the explicit inclusion of new components to simulation models. Can now explicitly model the effects of factors such as overtime rates and other working conditions on nurse exit/turnover rates. Allows policy makers a wider variety of policy scenarios to evaluate impact. Any reproduction requires permission from TOS co-PIs

51

Any reproduction requires permission from TOS co-PIs

52

1.

Average turnover rate is close to 20% per year in Canada, with the highest level in medical, surgical, combined medicalsurgical and ICUs.  For every 10 nurse vacancies the cost is 250K.  The highest turnover costs are attributed to temporary replacements and decrease in initial productivity of new hires.  Takes a new hire on average almost 8 weeks minimum to reach 100% role implementation; strategies need to be in place to support the transition.  Given the high cost associated with temporary replacement, leaders ought to reassess the costbenefit efficiency of employing temporary replacement staff over retention strategies.

Any reproduction requires permission from TOS co-PIs

53

Turnover is a system issue and is associated with decrease in job satisfaction, increase in likelihood of medical errors, overtime and environmental complexity.

2.

Better leadership is associated with better mental health, higher job satisfaction but decreased physical health status in nurses.

3.



Nurse leaders should ensure appropriate staffing resources that consider the needs of patients, and role responsibilities and health of nurses.

Any reproduction requires permission from TOS co-PIs

54

Better leadership is associated with higher productivity on the unit

4.

Evidence shows non-supportive working environments and poor relationships with nurse managers and other team members are contributing factors in nurses’ decision to leave.

5.

Role ambiguity and conflict on the units are associated with higher turnover rate for nurses.

6.

•

When nurses encounter competing demands or are unclear about their expectations, job stress and deterioration in mental health result.

Any reproduction requires permission from TOS co-PIs

55

Errors in patient care are related to higher levels of turnover and role ambiguity for nurses on the unit.

6.

•

7.

8.

Medical errors are 38% more likely to occur for each additional 10% increase in the turnover rate.

Units with higher proportion of overtime find it difficult to attract more experienced nurses; but these experienced nurses are crucial to supporting patient safety. Higher proportion of full time nurses is associated with lower nursing turnover.

Any reproduction requires permission from TOS co-PIs

56

Any reproduction requires permission from TOS co-PIs

57

1.

Effective leadership within healthcare organizations is pivotal to reducing nurse turnover because it addresses predisposing factors of turnover intent.  This study emphasizes the importance of building leadership capacity at all levels of organizations including practicing nurses, nurse managers, and clinical educators.  This type of leadership is important to promote and sustain a healthy work environment.

Any reproduction requires permission from TOS co-PIs

58

2.

This study indicates that the turnover of nurses is a major problem in Canadian Hospitals.  The mean turnover rate of 19.9% found in this work is a system issue that requires immediate action.  An average cost of $25,000 associated with nurse turnover is of concern to the system.  The key drivers of the cost associated with turnover are temporary replacement and overtime costs and initial decreased productivity of new hires.

Any reproduction requires permission from TOS co-PIs

59

3.

Turnover rates must be considered in all sectors and all types of delivery systems. ◦ The variables that contribute to turnover must be clearly defined and measured across sectors.  i.e. minimum data set.

◦ Overall within organizations and unit specific turnover rates must be monitored on an ongoing basis and strategies implemented to reduce the negative impact of turnover in organizations and at the unit level on patients, providers and systems, such as decreased continuity of care and failure to rescue among others.

Any reproduction requires permission from TOS co-PIs

60

4.

With implementation of innovative care delivery models, adequate resources are needed to address nurse turnover.  Appropriate skill mix, role clarity for team members and effective communication across caregiver groups is important.  An understanding of role expectations of team member may reduce role ambiguity and role conflict.  Resources should be in place and tools made available to optimize competencies of providers and facilitate maximization of scopes of practice.

Any reproduction requires permission from TOS co-PIs

61

5.

Policy options must be informed by evidence.

 Good quality, comparable and readily accessible data is needed.  Minimum Data Set  Linking data from a broad range of areas in organizations requires a sustainable investment in data and in capacity building.

Any reproduction requires permission from TOS co-PIs

62

To reduce nursing turnover, to enhance the quality of care and to improve nurse’ health and job satisfaction.

Any reproduction requires permission from TOS co-PIs

63

1. The evidence underscores the need for sustained investment in effective leadership and innovative model of care, where nurses are empowered in their work environment and recognized as a contributing asset to the delivery care system.

2. Optimize turnover rate and its cost to make turnover manageable and efficient.

Any reproduction requires permission from TOS co-PIs

64

3. Implement and evaluate the impact of models of care which reduce competing resources and demands on frontline nurses to reduce turnover, promote patient safety and nurse satisfaction in the delivery care system.

4. Invest in the linkage and access of data to standardize measures and indicators on turnover. Also, an ongoing investment in data is needed to better understand the needs of people and the work and productivity of nurses.

Any reproduction requires permission from TOS co-PIs

65



In summary, partnerships and a variety of strategies involving -–ministries of health, funders, regulatory bodies, decision makers, nurse managers and clinical educators, and frontline healthcare providers -- is needed to address the cost of turnover and to alleviate or address the factors that influence turnover

Any reproduction requires permission from TOS co-PIs

66

Any reproduction requires permission from TOS co-PIs

68